Well-defined transition metal carbene complexes have emerged as the catalysts of choice for a wide variety of selective olefin metathesis transformations [F. Z. Dxc3x6brwald, Metal Carbenes in Organic Synthesis; Wiley VCH, Weinheim, 1999]. These transformations include olefin cross metathesis (CM), ring-opening metathesis (ROM), ring-opening metathesis polymerization (ROMP), ring-closing metathesis (RCM), and acyclic diene metathesis (ADMET) polymerization [K. J. Ivin and J. C. Mol, Olefin Metathesis and Metathesis Polymerization; Academic Press, London, 1997]. Of particular importance has been the development of ruthenium carbene catalysts demonstrating high activity combined with unprecedented functional group tolerance [T. M. Trnka and R. H. Grubbs, Acc. Chem. Res., 2001, 34, 18-29]. Olefin metathesis serves as a key reaction for the development of a range of regioselective and stereoselective processes. These processes are important steps in the chemical synthesis of complex organic compounds and polymers and are becoming increasingly important in industrial applications. [see for example Pederson and Grubbs U.S. Pat. No. 6,215,019].
An initial concern about using ruthenium olefin metathesis catalysts in commercial applications has been reactivity and catalyst lifetime. The original breakthrough ruthenium catalysts were primarily bisphosphine complexes of the general formula (PR3)2(X)2Ruxe2x95x90CHRxe2x80x2 wherein X represents a halogen (e.g., Cl, Br, or I), R represents an alkyl, cycloalkyl, or aryl group (e.g., butyl, cyclohexyl, or phenyl), and Rxe2x80x2 represents an alkyl, alkenyl, or aryl group (e.g., methyl, CHxe2x95x90CMe2, phenyl, etc.). Examples of these types of catalysts are described in U.S. Pat. Nos. 5,312,940, 5,969,170 and 6,111,121. Though they enabled a considerable number of novel transformations to be accomplished, these bisphosphine catalysts can exhibit lower activity than desired and, under certain conditions, can have limited lifetimes.
More recent developments of metathesis catalysts bearing a bulky imnidizolylidine ligand [Scholl et. al. Organic Letters 1999, 1, 953-956] such as 1,3-dimesitylimidazole-2-ylidenes (IMES) and 1,3-dimesityl4,5-dihydroimidazol-2-ylidenes (sIMES), in place of one of the phosphine ligands have led to greatly increased activity and stability. For example, unlike prior bisphosphine complexes, the various imidizolyidine catalysts effect the efficient formation of trisubstituted and tetrasubstituted olefins through catalytic metathesis. Examples of these types of catalysts are described in PCT publications WO 99/51344 and WO 00/71554. Further examples of the synthesis and reactivity of some of these active ruthenium complexes are reported by A. Fxc3xcrstner, L. Ackermann, B. Gabor, R. Goddard, C. W. Lehmann, R. Mynott, F. Stelzer, and O. R. Theil, Chem. Eur. J., 2001, 7, No. 15, 3236-3253; S. B. Gaber, J. S. Kingsbury, B. L. Gray, and A. H. Hoveyda, J. Am. Chem. Soc., 2000, 122, 8168-8179; Blackwell H. E., O""Leary D. J., Chatterjee A. K., Washenfelder R. A., Bussmann D. A., Grubbs R. H. J. Am. Chem. Soc. 2000, 122, 58-71; Chatterjee, A. K., Morgan J. P., Scholl M., Grubbs R. H. J. Am. Chem. Soc. 2000, 122, 3783-3784; Kingsbury, J. S.; Harrity, J. P. A.; Bonitatebus, P. J.; Hoveyda, A. H. J. Am. Chem. Soc. 1999, 121, 791-799; Harrity, J. P. A.; Visser, M. S.; Gleason, J. D.; Hoveyda, A. H. J. Am Chem. Soc. 1997, 119, 1488-1489; and Harrity, J. P. A.; La, D. S.; Cefalo, D. R.; Visser, M. S.; Hoveyda, A. H. J. Am. Chem. Soc. 1998, 120, 2343-2351.
The improvements in catalyst activity and expansion of potential substrates resulted in the ruthenium metathesis systems becoming attractive candidates for use in industrial scale processes. In particular, many of the targeted products of olefin metathesis are useful as intermediates in flavors and fragrances, pharmaceuticals and other fine chemicals. Thus, a second major concern has involved ruthenium residues that may be present in the products produced by metathesis. To address this issue, several catalyst removal techniques have been developed [Maynard and Grubbs in Tetrahedron Letters 1999, 40, 4137-4140; L. A. Paquette, J. D. Schloss, I. Efremov, F. Fabris, F. Gallou, J. Mendez-Andino and J. Yang in Org. Letters 2000, 2,1259-1261; and Y. M. Ahn; K. Yang, and G. I. Georg in Org. Letters 2001, 3, 1411], including that described by Pederson and Grubbs [Pederson and Grubbs, U.S. Pat. No. 6,215,049] which is still the most amenable to large scale reactions. Ruthenium metathesis catalysts with a wide range of reactivity and that could be easily removed from the product were now available.
Further progress towards catalyst selectivity, stability, and removal has been recently published by Hoveyda and others [Kingsbury, J. S.; Harrity, J. P. A.; Bonitatebus, P. J.; Hoveyda, A. H. J. Am. Chem. Soc. 1999, 121, 791-799] with the demonstration of new, readily recyclable catalyst systems containing chelating carbene species (FIG. 1) that are exceptionally stable and can even be purified by column chromatography in air. For example, the tricyclohexylphosphine-ligated variant, Catalyst 601 (FIG. 1), can be recovered in high yield from the reaction mixture by simple filtration through silica. Hoveyda and coworkers also demonstrated [Cossy, J.; BouzBouz, S.; Hoveyda, A. H. J. Organometallic Chemistry 2001, 624, 327-332] that by replacement of the phosphine with the sIMES ligand, Catalyst 627 (FIG. 1) actively promotes the cross-metathesis of acrylonitrile and terminal olefins in moderate to excellent yields (20% to 91%) with a cis to trans olefin ratios that range from 2:1 to over 9:1. Related chelating carbene catalysts are described in U.S. Patent Application Publication No. 2002/010713 8 and U.S. Pat. No. 6,306,987. 
Prior methods used to make these chelating carbene complexes include treating (Ph3P)3RuCl2 with the appropriate diazo species at low temperatures or treatment of a metathesis-active metal carbene complex with the parent styrene in the presence of CuCl followed by column chromatography FIG. 2). While both of these methods yield the desired compound, they are difficult to scale up. Maintaining very low temperatures on larger reaction vessels requires expensive equipment, and diazo species are prone to violent decomposition under certain conditions. Using the o-isopropoxy styrene/CuCl route is also not amenable to large scale due to the requirement to purify the product by column chromatography. A further shortcoming includes the use of the Wittig reaction to yield the key styrene intermediate. Wittig reactions are not convenient on a commercial scale because of the high costs of the reagents and the byproduct, triphenylphosphine oxide, produces an excessive mass of waste. Alternatives to Wittig reactions would include Heck, Stille or Suzuki coupling of vinyl trialkyltin, vinyl triflates or vinyl borate; respectively, to a halo-phenol substrate. These starting materials are generally expensive, and the reactions with trialkyl tin reagents involve toxic compounds which require special waste disposal procedures. Finally the styrene itself is prone to polymerization under some of the conditions required to make the xe2x80x9cHoveyda-typexe2x80x9d catalysts. Therefore, there is a need for an efficient and economical synthesis to chelating carbene type ruthenium metathesis catalysts in larger quantities. 
The present invention describes efficient and versatile routes to useful and valuable Hoveyda-type catalysts with chelating phenyl carbene ligands while eliminating expensive and toxic reagents. The present invention describes the synthesis of substituted olefins that are precursors to catalyst complexes and their use as reagents to prepare olefin metathesis catalysts with chelating carbene ligands.
The present invention comprises methods for the use of novel chelating ligand precursors for the preparation of olefin metathesis catalysts. The resulting catalysts comprise monomeric species which are air stable, are capable of promoting various forms of metathesis reactions in a highly efficient manner, and can be recovered from the reaction mixture and reused.
One embodiment of the present invention is the use of internal olefin compounds, specifically beta-substituted styrenes, as ligand precursors instead of terminal olefin compounds such as unsubstituted styrenes (FIG. 3). Although internal olefins tend to be less reactive than terminal olefins, we have surprisingly found that the beta-substituted styrenes are sufficiently reactive to efficiently produce the desired catalyst complexes. Compared with the styrene compounds, the beta-substituted styrenes are much easier and less costly to prepare in large quantities and are more stable in storage and use since they are less prone than terminal styrenes to spontaneous polymerization. 
Another embodiment of the present invention are methods of preparing chelating-carbene metathesis catalysts without the use of CuCl as previously required. In previous reports, CuCl was used to sequester phosphine ligands which shifts the equilibrium of metathesis reactions to product formation. The use of CuCl in large scale synthesis is problematic in that the resulting metathesis catalyst must be purified by chromatography before recrystallization, requiring large volumes of silica and solvent [Kingsbury et. al. J. Am. Chem. Soc. 1999, 121, 791-799]. The present invention eliminates the need for CuCl by replacing it with organic acids, mineral acids, mild oxidants or even water, resulting in high yields of Hoveyda-type metathesis catalysts. The phosphine byproduct can be removed by an aqueous wash or filtration, thereby eliminating the chromatography step and allowing catalysts to be readily isolated by crystallization from common organic solvents.
A further embodiment of the present invention is an efficient method for preparing chelating-carbene metathesis catalysts by reacting a suitable ruthenium complex in high concentrations of the novel ligand precursors followed by crystallization from an organic solvent. For example, in this manner Catalyst 601 can be simply isolated by filtering a hexane solution of the reaction mixture resulting from the reaction of neat ligand precursor and a ruthenium carbene complex. By using the beta-substituted styrene derivatives, the excess, unreacted ligand is recoverable from such reaction mixtures and can be reused. This is difficult with the parent styrenes due to the propensity of those materials to polymerize under reaction and workup conditions.
The present invention describes the synthesis of xe2x80x9cHoveyda-typexe2x80x9d chelating carbene metathesis catalysts from the cross metathesis of novel ligand precursors and metal carbene complexes. Although any metathesis-active metal carbene complex is suitable for use in the present invention, preferred metal complexes include the Grubbs-type compounds described in U.S. Pat. Nos. 5,312,940, 5,969,170, 6,077,805, 6,111,121 and 6,426,419 and PCT publications WO 99/51344 and WO 00/71554. These complexes have the general formula X1X2L1(L2)mMxe2x95x90CR1R1, wherein X1 and X2 are each, independently, any anionic ligand; L1 and L2 are each, independently, any neutral electron donor ligand; m is 1 or 2; M is ruthenium or osmium; and R1 and R2 are each, independently, hydrogen or a group selected from the group consisting of alkyl, alkenyl, alkynyl, aryl, alkoxy, alkenyloxy, alkynyloxy, aryloxy, alkoxycarbonyl, alkylthio, alkylsulfonyl, alkylsulfinyl and trialkylsilyl, any of which may be optionally substituted with a functional group selected from the group consisting of halogen, alkyl, alkenyl, alkynyl, aryl, heteroaryl, alkoxy, alkenyloxy, aryloxy, alkoxycarbonyl, carbonyl, alkylamino, alkylthio, alkylsulfonyl, nitrile, nitro, alkylsulfinyl, trihaloalkyl, perfluoroalkyl, carboxylic acid, ketone, aldehyde, nitrate, cyano, isocyanate, hydroxyl, ester, ether, amine, imine, amide, sulfide, disulfide, sulfonate, carbamate, silane, siloxane, phosphine, phosphate, or borate. In these preferred metal carbene complexes, R1 and R2 may be linked to form a cyclic group, and any two or three of X1, X1, and L1 may be linked to form a multidentate ligand and two L2 ligands, if m=2, may be linked to form a bidentate ligand. One type of chelating carbene complex, for example catalyst 601 or catalyst 627, may also be reacted with the ligand precursors of the present invention to make different chelating carbene complexes.
The ligand precursors of the present invention are functionalized beta-substituted styrene compounds, which may be conveniently prepared by the isomerization of functionalized allylbenzenes, with the structure shown in FIG. 4. 
Wherein:
Y is a heteroatom such as oxygen (O), sulfur (S), nitrogen (N), or phosphorus (P);
Z is a group selected from hydrogen, alkyl, aryl, functionalized alkyl, functionalized aryl where the functional group(s) may independently be one or more or the following: alkoxy, aryloxy, halogen, carboxylic acid, ketone, aldehyde, nitrate, cyano, isocyanate, hydroxyl, ester, ether, amine, imine, amide, sulfide, disulfide, carbamate, silane, siloxane, phosphine, phosphate, or borate.
n is 1, in the case of a divalent heteroatom such as O or S, or 2, in the case of a trivalent heteroatom such as N or P;
R3 and R4 are independently selected from the group consisting of hydrogen, C1-C20 alkyl, C6-C20 aryl, C1-C20 alkoxy or C6-C20 aryloxy;
R5, R6, R7, and R8 are each, independently, selected from the group consisting of hydrogen, halogen, alkyl, alkenyl, alkynyl, aryl, heteroaryl, alkoxy, alkenyloxy, aryloxy, alkoxycarbonyl, carbonyl, alkylamino, alkylthio, alkylsulfonyl, nitrile, nitro, alkylsulfinyl, trihaloalkyl, perfluoroalkyl, carboxylic acid, ketone, aldehyde, nitrate, cyano, isocyanate, hydroxyl, ester, ether, amine, imine, amide, sulfide, disulfide, sulfonate, carbamate, silane, siloxane, phosphine, phosphate, or borate. Additionally, any two or more of R5, R6, R7, and/or R8 may be independently connected through hydrocarbon or functionalized hydrocarbon groups forming aliphatic or aromatic rings. Furthermore, one who is skilled in the art will recognize that R8 should be chosen such that its steric bulk or chemical functionality does not interfere with the cross-metathesis reaction between the ligand precursor and the metal carbene complex. Any one or more of R5, R6, R7 and R8 (but preferably any of R5, R6 and R7) may be a linker to a solid support such as silica, swellable polymeric resins, dendritic polymers, and the like as, for example, described in U.S. Patent Application Publication No. 2002/0107138 or by Grela (et al.) in Tetrahedron Letters, 2002, 43, 9055-9059 for terminal-styrene ligand precursors.
Preferred ligand precursors are beta-methyl styrenes wherein Y is oxygen or sulfur; n is 1; Z is alkyl, aryl or trialkylsilyl; and R3 and R4 are both hydrogen. Particularly preferred ligand precursors are alkoxy-substituted beta-methyl styrenes wherein Y is oxygen; n is 1; Z is methyl, isopropyl, sec-butyl, t-butyl, neopentyl, benzyl, phenyl or trimethylsilyl; and R3, R4, R5, R6, R7 and R8 are all hydrogen. Examples of particularly preferred ligand precursors of these types include 2-methoxy-xcex2-methylstyrene, 2-isopropoxy-xcex2-methylstyrene and 2-isopropoxy-3-phenyl-xcex2-methylstyrene: 
The precursor compounds for chelating ligands are easily prepared in high yields from commercially available starting materials. Treatment of allyl aryl compounds with an isomerization catalyst cleanly migrates the double bond one carbon closer to the aryl ring forming a beta-substituted styrenic olefin (FIG. 4). We have found that (PPh3)3RuCl2 is a preferred, highly active isomerization catalyst that is effective at amounts ranging from about 0.001 to 20 mole percent relative to the allyl aryl compound. It is preferable to isomerize the allylphenol compounds prior to further fanctionalization, since the hydroxy protons serve to activate the catalyst and the reactions can therefore be run neat. For other compounds without their own protic source, it is necessary to add an alcohol or other proton source to initiate the isomerization catalysis. From the structures shown in FIG. 4, one skilled in the art can appreciate the diversity of substitution on the aromatic system that can be achieved. This allows the ligand to be fine-tuned for specific applications. For the case where Y is oxygen, a wide variety of allyl phenol starting materials are easily produced by the Claisen rearrangement (FIG. 5) of allylic aryl ethers [March""s Advanced Organic Chemistry; 5th Edition, Eds. M. B. Smith and J. March; John Wiley and Sons, New York, N.Y. 2001, pp. 1449-1452]. Similar rearrangements are operative for the case where Y is nitrogen, although more forcing conditions are typically required. 
The above described ligand precursors can be used to prepare metathesis catalysts with a chelating carbene group. The preferred chelating carbene complexes have the structure shown in FIG. 6. The most preferred chelating carbene complexes that are made by this method are Hoveyda-type complexes including, but not limited to, catalyst 601 and catalyst 627. In the most basic practice of the present invention, as with the parent styrenes, it is possible to mix a metathesis active metal carbene complex with the ligand precursor in a suitable solvent to effect the transformation. Preferred solvents typically include, but are not limited to, chlorinated solvents (such as methylene chloride, dichloroethane, chlorobenzene, and dichlorobenzenes), ethereal solvents (such as tetrahydrofuran or dioxane), aromatic solvents (such as benzene, toluene, or xylenes), and hydrocarbon solvents (such as hexanes, heptane, and petroleum distillate fractions). In general, at least one equivalent, and preferably an excess amount, of the ligand precursor is utilized. Depending upon the reactivity of the metathesis-active metal carbene complex, the reaction may proceed at room temperature, or even lower, or may need to be heated. As the progress of these reactions can be conveniently monitored by a variety of techniques including thin-layer chromatography (TLC), those skilled in the art can readily ascertain the appropriate conditions of time and temperature to achieve high conversions to the desired chelating carbene complexes. 
In general, these reactions proceed more slowly and/or require somewhat higher reaction temperatures than comparable reactions with terminal styrenes. In order to increase the reaction rates and achieve higher conversion, high ratios of ligand precursor to metal carbene complex can be utilized In fact, in the practice of the present invention, the reaction can be performed using neat ligand precursor as the solvent. In general, three to ten mole-equivalents of ligand precursor will give reasonable reaction rates and high conversions, although higher amounts may be used. This approach cannot be utilized with the terminal styrene ligand precursors due to their propensity to spontaneously polymerize under the reaction conditions. Upon completion of the reaction, the ligand precursor can be distilled off of the reaction mixture and the chelating carbene product recrystallized from an appropriate solvent Alternatively, the chelating carbene product can be precipitated by the addition of an appropriate nonsolvent and the unreacted ligand precursor recovered by distillation of the mother liquor. The beta-substituted styrene compounds are sufficiently robust so that high recoveries can be achieved by these methodologies, which would not be practical with the easily polymerized terminal styrenes.
In general, treatment of one mass equivalent of a ruthenium carbene complex with between 1 and 20 mass equivalents of ligand precursor in the presence of an optional co-solvent (generally between about 1-20 mass equivalents relative to the ruthenium complex) yields a thick mixture that gradually looses viscosity during the course of the reaction. Optionally the mixture can be heated or cooled. The mixture may also be exposed to a static or dynamic vacuum. The reaction is preferably conducted under an inert atmosphere but may be conducted in air unless the metal carbene complex is particularly air-sensitive. After 3 hours to 7 days of stirring, the reaction is usually complete and the product may be isolated as described above.
A complementary method for increasing reaction rates and conversion utilizes an additive to sequester the ligand that is displaced from the metal carbene complex during the course of the reaction. When the displaced ligand is a phosphine ligand, as is typical, the sequestering agent that has been commonly used in cuprous cloride (CuCl), although this is difficult to separate from the product without using chromatograhy which is impractical at large scale. Surprisingly, we have found that replacement of the CuCl with mineral acids, organic acids or mild oxidants in the presence of the ligand precursors of the present invention is also very effective. Treatment of ruthenium carbene complexes with between 1 to 10 equivalents of ligand precursor and between 0.1 to 10 equivalents acid or mild oxidant yields the new catalyst containing the chelating carbene moity. After the reaction is complete, the displaced ligand and the sequestering agent can be readily removed from the mixture by extraction into water. The product can then be simply crystallized from the resulting solution in organic solvents in very high yield, eliminating the need for column chromatography. Preferred sequestering agents include hydrochloric acid, solutions of hydrogen chloride in ethereal solvents (such as diethyl ether, tetrahydrofuran, or dioxane), gaseous hydrogen chloride dissolved in the reaction mixture, glacial acetic acid, bleach, and dissolved oxygen. Water can be utilized as a sequestering agent for particularly basic ligands such as tricyclohexylphosphine (TCBP or PCy3). The use of sequestering agents is particularly perferred when using very robust metal carbene complexes such as those containing IMES or sIMES ligands. When using less robust complexes such as ruthenium carbenes ligated with two phosphine ligands, greater care is needed and it is desirable to utilize the mildest sequestering agents or to slowly add the sequestering agents over the course of the reaction.